Article Figures & Tables
Tables
Risk factor; study Age Study design No. of studies in review Total no. of participants Summary of results Ganz et al. (2) Systematic review of prospective cohort studies 18 (3 studies of age) 19 178 For first and second studies combined, probability of falls by age category: 65–74 yr, 31%–32%; 70–74 yr, 22%–33%; 75–79 yr, 25%–36%; ≥80 yr, 34%–37%
For third study, increased risk of falling at least once in next 11 mo among older patients (OR per age category 1.90; p < 0.001)
Probability of falls, by age category: 65–69 yr, 14%; 70–74 yr, 16%; 75–79 yr, 24%; ≥ 80 yr, 34%
Previous falls Ganz et al. (2) Systematic review of prospective cohort studies 18 (11 studies of history of falls, 4 with extractable data) 19 178 Occurrence of fall in past year associated with subsequent falls (LR range 2.3–3.8) Cognitive impairment Ganz et al. (2) Systematic review of prospective cohort studies 18 (8 studies of cognitive impairment, 2 with extractable data) 19 178 In first study, presence of ≥ 5 errors on Short Portable Mental Status Questionnaire associated with ≥ 1 falls (LR 4.2, 95% CI 1.9–9.6)
In second study, history of dementia associated with ≥ 1 falls (LR 17, 95% CI 1.9–149) and with ≥ 2 falls (LR 13, 95% CI 2.3–79)
Visual impairment Ganz et al. (2) Systematic review of prospective cohort studies 18 (11 studies of visual impairment, 3 studies showed significant results) 19 178 For first and second studies, visual impairment increased the likelihood of a fall (OR 1.6 and 2.0, respectively
For third study, using Bailey–Lovie charts to measure visual acuity, each additional letter read correctly off the chart at baseline was associated with lower risk of falls (OR 0.96)
Medications Woolcott et al. (10) Meta-analysis of cohort, cross-sectional, case–control studies 22 79 081 Antidepressants, OR 1.68 (95% CI 1.47–1.91)
Antihypertensive agents, OR 1.24 (95% CI 1.01–1.50)
Benzodiazepines, OR 1.57 (95% CI 1.43–1.72)
Diuretics, OR 1.07 (95% CI 1.01–1.14)
Neuroleptics and antipsychotics, OR 1.59 (95% CI 1.37–1.83)
NSAIDs, OR 1.21 (95% CI 1.01–1.44)
Sedatives and hypnotics, OR 1.47 (95% CI 1.35–1.62)
Antidepressants, OR 1.36 (95% CI 1.13–1.76)
Benzodiazepines, OR 1.41 (95% CI 1.20–1.71
Hegeman et al. (11) Systematic review of case–control, prospective and cross-sectional studies 13 (12 with extractable OR values) 209 015 NSAIDs, OR range 1.13 (95% CI 0.93–1.38) to 4.35 (95% CI 1.79–10.91) Sterke et al. (12) Systematic review of prospective cohort studies 17 61 392 Large range of OR and RR values for various combinations of psychoactive medications (antidepressants, tranquilizers, antipsychotics, benzodiazepines) used for patients with dementia living in nursing homes: From 17 studies, OR range 1.13–5.67 and RR range 1.32–10.3
Multiple drugs (3/3 studies: n = 177–282, RR range 1.30–10.3), antidepressants (10/12 studies: n = 78–2428, range of effect sizes 1.1–7.6) and antianxiety drugs (2/2 studies: n = 2015, RR 1.32 and n = 18 855, OR 1.22) increase fall risk
Evidence for other psychoactive drug classes is limited (antipsychotics), inconclusive (sedatives) or no association (hypnotics)
Kim et al. (13) Meta-analysis of RCTs 40 total studies 22 studies of cholinesterase inhibitors in analysis of falls 9882 (for cholinesterase inhibitors) Cholinesterase inhibitors: syncope, OR 1.53 (95% CI 1.02–2.30) 14 studies of memantine, with 13 in analysis of falls 3583 (for memantine) Memantine: fewer fractures, OR 0.21 (95% CI 0.05–0.85) Functional limitations, disabilities in ADLs Bloch et al. (14) Meta-analysis of RCTs and observational studies (cohort, cross-sectional, case–control) 177 19 178 Any impairment in ADLs, OR 2.26 (95% CI 2.09–2.45)
Any impairment in instrumental ADLs, OR 2.10 (95% CI 1.68– 2.64)
Ganz et al. (2) Systematic review of prospective cohort studies 18 (10 studies specifically on impairment of ADLs, 3 studies showed significant results) 19 178 2 studies reported LR values: In first study, inability to rise from chair of knee height without using chair arms was associated with increased risk of ≥ 1 falls among men (LR 4.3, 95% CI 2.3–7.9); not significant in women
In second study, presence of ≥ 5 of 11 physical impairments (mostly ADLs) was associated with increased risk of ≥ 1 falls (LR 1.9, 95% CI 1.4–2.6)
Home hazards Letts et al. (15) Meta-analysis of cohort and cross-sectional studies 100 25 145 Various indoor and outdoor home hazards (e.g., bathroom, environmental) were associated with increased risk of falls (OR 1.15, 95% CI 0.97–1.36)
Use of mobility aids was associated with significantly increased risk of falls in community (OR 2.07, 95% CI 1.59–2.71) and institutional settings (OR 1.77, 95% CI 1.66–1.89)
Orthostatic hypotension Ganz et al. (2) Systematic review of prospective cohort studies 18 (4 studies specific to orthostatic hypotension) 19 178 In 4 studies, no association between orthostatic hypotension and falls when other risk factors were considered
In 1 study, increase in pulse (< 6 beats/min) measured 30 s after standing up was weakly associated with falls (LR 1.4, 95% CI 1.0–1.9)
Balance impairment Muir et al. (16) Meta-analysis of prospective studies 23 60 602 Overall fall risk, OR 1.98 (95% CI 1.60–2.46) Impairment of gait or balance Ganz et al. (2) Systematic review of prospective cohort studies 18 (15 studies on impairment of gait or balance) 19 178 Of 15 studies found, 4 reported LR for clinically detected abnormality of gait or balance (LR range 1.7–2.4)
For presence of lower-extremity disability (i.e., reported problem with strength, sensation or balance), LR 1.8 (95% CI 1.5–2.2)
Note: ADLs = activities of daily living, CI = confidence interval, LR = likelihood ratio, NSAID = nonsteroidal anti-inflammatory drug, OR = odds ratio, RCT = randomized controlled trial, RR = relative risk.
- Table 2:
Studies evaluating interventions for preventing falls among older people, by setting
Setting; study Study design No. of studies in review Total no. of participants Intervention Outcome Community Thomas et al. (20) Systematic review of RCTs and controlled trials with masked assessment of outcome 7 1 503 Otago Exercise Programme v. usual care or social visits Lower risk of death (relative risk 0.45, 95% CI 0.25 to 0.80)
Lower fall rate (incidence rate ratio 0.68, 95% CI 0.56 to 0.79)
Risk of serious injury from fall (relative risk 1.05, 95% CI 0.91 to 1.22)
Clemson et al. (21) Meta-analysis of RCTs 6 3 298 Home assessment and modification, including hazard reduction, behavioural changes, footwear, ADLs, instrumental ADLs, mobility, home visits, home modifications, vision assessment Lower risk of falls associated with home assessment (relative risk 0.79, 95% CI 0.65 to 0.97) Gillespie et al. (22) Systematic review of RCTs and quasi-RCTs 111 55 303 Exercise interventions combined with 13 other approaches, including education, home safety intervention, and supplementation with vitamin D and calcium Multicomponent group exercise associated with reduced rate of falls and risk of falling (rate ratio 0.78, 95% CI 0.71 to 0.86; relative risk 0.83, 95% CI 0.72 to 0.97)
Tai chi associated with reduced risk of falling (rate ratio 0.63, 95% CI 0.52 to 0.78; relative risk 0.65, 95% CI 0.51 to 0.82)
Individually prescribed multicomponent home-based exercise associated with reduced risk of falling (rate ratio 0.66, 95% CI 0.53 to 0.82; relative risk 0.77, 95% CI 0.61 to 0.97)
Assessment and multifactorial intervention associated with reduced rate of falls (rate ratio 0.75, 95% CI 0.65 to 0.86)
Vitamin D not associated with reduced rate of falls (rate ratio 0.95, 95% CI 0.80 to 1.14; relative risk 0.96, 95% CI 0.92 to 1.01)
Gillespie et al. (23) Systematic review of RCTs and quasi-RCTs 159 79 193 Numerous approaches, including single interventions (59 trials) and multifactorial approaches (40 trials) Group and home-based exercise programs and home safety interventions associated with reduced rate of falls and risk of falling
Multifactorial assessment and intervention programs associated with reduced rate of falls but no reduction in risk of falling
Tai chi associated with reduced risk of falling
Vitamin D supplementation did not appear to reduce falls but may be effective in people with lower vitamin D levels before treatment
See Table 3 for further details
Zijlstra et al. (24) Systematic review of RCTs 19 Unavailable Home-based multifactorial programs and single interventions (i.e., tai chi, exercise, hip protector intervention) Reduction in fear of falling among older adults living in the community Orr et al. (25) Systematic review of RCTs 29 2 174 Progressive resistive training v. usual daily activity, usual care or activities that enhance blinding of intervention Intervention needs further evaluation; limited evidence that progressive resistive training in isolation improves balance Michael et al. (26) Systematic review of RCTs 47 152 Various primary care interventions to prevent falls in people at higher risk of falling, including comprehensive multifactorial assessment and management, exercise and physical therapy interventions, and vitamin D supplementation Reduced risk of falling with multifactorial assessments and provision of medical and social care (relative risk 0.75, 95% CI 0.58 to 0.99)
Reduced risk of falling with vitamin D supplementation (pooled relative risk 0.83, 95% CI 0.75 to 0.91)
Reduced risk of falling with exercise and physical therapy (pooled relative risk 0.87, 95% CI 0.81 to 0.94)
No reduction in risk of falling with noncomprehensive multifactorial assessment and referral or limited management (relative risk 1.04, 95% CI 0.98 to 1.10)
No significant finding for hip protectors (relative risk 0.89, 95% CI 0.75 to 1.06), clinical education and vision correction
Campbell and Robertson (27) Meta-regression of RCTs 14 5 968 Single and multifactorial interventions** For populations at risk, targeted single interventions were as effective as multifactorial interventions; possibly more acceptable and more cost-effective Reduction of falls similar with single and multicomponent interventions: Single component, pooled rate ratio 0.77 (95% CI 0.67 to 0.89)
Multicomponent, rate ratio 0.78 (95% CI 0.68 to 0.89)
Nursing care facilities Cameron et al. (28),* Systematic review of RCTs 41 25 422 Multifactorial interventions (combinations of exercise, medications, environmental modification, knowledge, and measures to address other factors such as incontinence, fluid intake, nutrition, psychological concerns, vitamin D levels) and single-factor interventions (e.g., medications, exercise, knowledge) Multifactorial interventions in hospital setting reduced rate of falls (rate ratio 0.69, 95% CI 0.49 to 0.96) and risk of falling (risk ratio 0.73, 95% CI 0.56 to 0.96)
Vitamin D supplementation effective in reducing rate of falls in nursing care facilities (rate ratio 0.72, 95% CI 0.55 to 0.95) but not risk of falling (risk ratio 0.98, 95% CI 0.89 to 1.09)
Supervised exercise appeared effective in subacute care hospital setting (risk ratio 0.44, 95% CI 0.20 to 0.97)
Inconsistent results for nursing care with supervised exercise intervention
Multifactorial intervention in nursing home: no significant reduction in rate of falls (rate ratio 0.82, 95% CI 0.62 to 1.08) or risk of falling (risk ratio 0.93, 95% CI 0.86 to 1.01), but post hoc analysis showed that multifactorial intervention reduced rate of falls (rate ratio 0.60, 95% CI 0.51 to 0.72) and risk of falling (risk ratio 0.85, 95% CI 0.77 to 0.95)
Cusimano et al. (29)† RCTs 5 2 395 Multifaceted programs (i.e., combinations of education, environmental modification, home assessments, review of drug regimen, exercise sessions and programs, personal educational consultation, gait aid maintenance, vision assessment, use of hip protectors) Some efficacy described for multifaceted programs. No combined data of the trials. Hospital, acute care de Morton et al. (30) Systematic review of RCTs and controlled clinical trials 9 4 223 Multidisciplinary interventions; only those that included exercise were compared with “usual hospital care” Small but significant increase in discharge home from hospital with multidisciplinary interventions (relative risk 1.08, 95% CI 1.03 to 1.14; number needed to treat 16, 95% CI 11 to 43) Mixed setting (community, institutions, acute care hospitals) Kalyani et al. (31) Meta-analysis of RCTs 10 2 932 Vitamin D (200–1000 IU daily) v. calcium or placebo Reduction in risk of falling (relative risk 0.86, 95% CI 0.79 to 0.93)
Post hoc analysis with 7 additional studies without explicit definition of falling (i.e., total 17 studies) yielded smaller benefit (relative risk 0.92, 95% CI 0.87 to 0.98), with significant intergroup differences favouring adjunctive calcium over no calcium
Bischoff-Ferrari et al. (32)‡ Meta-analysis of RCTs 8†† 2 426 Vitamin D (700–1000 IU daily) Reduction in risk of falling (pooled relative risk 0.81, 95% CI 0.71 to 0.92)
Achieved serum 25(OH)D concentrations of 60 nmol/L or more (pooled relative risk 0.77, 95% CI 0.65 to 0.90)
Active forms of Vitamin D (pooled relative risk 0.78, 95% CI 0.64 to 0.94)
Murad et al. (33) Meta-analysis of RCTs 26 45 782 Vitamin D (200–1000 IU daily) Reduction in risk of at least 1 fall (OR 0.86, 95% CI 0.77 to 0.96), not significant in vitamin D supplementation without co-administration of calcium Low et al. (34)§ Meta-analysis of RCTs 7 1 972 Tai chi as single intervention Potential reduction in fall rate or risk of falls among older adults; no pooled data Sitjà-Rabert et al. (35)§ Meta-analysis of RCTs 16‡‡ 957 Whole-body vibration programs Improved isometric strength of knee muscle 18.30 Nm (95% CI 7.95 to 28.65), muscle power 10.44 W (95% CI 2.85 to 18.03)
No significant difference for comparison with conventional exercise, but significant difference for comparison with control group
Gates et al. (36)¶ Meta-analysis of RCTs and quasi-RCTs 19 6 397 Multifactorial fall prevention programs§§ Decrease in number of people who fell (combined relative risk 0.91, 95% CI 0.82 to 1.02)
No reduction in fall-related injuries (combined relative risk 0.90, 95% CI 0.68 to 1.20)
Vaapio et al. (37) Systematic review of RCTs 12 2 357 Fall prevention intervention with assessment of quality of life Positive effect on quality of life in only a few studies
Quality of life domains included physical function, social function, vitality, mental health, environmental domain
No pooled data
Fairhall et al. (38)§ Systematic review and meta-analysis of RCTs 19 3 616 Exercise interventions, with measurement of participation in intervention and effect on participation in life roles Pooled estimate of effect of exercise on participation: (Hedges’ g 0.16 (95% CI 0.04 to 0.27, p = 0.006)¶¶
Point estimate of effect of multifaceted intervention with exercise component on participation: Hedges’ g 0.25 (95% CI −0.04 to 0.53, p = 0.09)
Effect of exercise as a single intervention: Hedges’ g 0.09 (95% CI −0.01 to 0.19, p = 0.07)
Note: ADLs = activities of daily living, CI = confidence interval, RCT = randomized controlled trial.
↵* Nursing care facilities and hospitals (acute and subacute [subacute defined as care provided to patients still in need of medical and nursing services]), analyzed separately.
↵† Residential care, including nursing homes.
↵‡ Community rehabilitation and acute care hospitals.
↵§ Community and long-term care settings.
↵¶ Emergency department, primary care and community settings.
↵** Includes various combinations of home assessments, comprehensive geriatric assessments, hospital-based medical assessments, diagnostic home visits, Stepping On program (group sessions for balance and strengthening exercises, home and community environmental and behavioural safety measures, encouragement of regular review of vision and medications, follow-up home visit by occupational therapist), specific exercise program (i.e., physiotherapy and occupational therapy, tai chi and balance training, Otago Exercise Programme), medication withdrawal v. social visits or no intervention or usual care.
↵†† 2 RCTs examined active forms of vitamin D.
↵‡‡ 16 studies included in the qualitative synthesis and 10 studies included in the quantitative synthesis (meta-analysis).
↵§§ Includes various combinations of occupational therapy, home assessment, risk assessment, geriatric assessment, gait and balance exam, cardiovascular assessment, drug review, vision, psychological assessment, personal care aids, self-management group session, counselling, motivational video, standardized and individualized fall prevention, other assessments including foot, positional transfer, function, lower limb disability, alcohol use, hearing v. “usual care” or no intervention.
↵¶¶ Hedges’ g calculates effect size and measures based on a standardized difference.
- Table 3:
Interventions for preventing falls among older people living in the community (Gillespie et al. (23))
Intervention Rate of falls* Risk of falling* Rate ratio (95% CI) No. of participants No. of trials Relative risk (95% CI) No. of participants No. of trials Exercise Tai chi 0.72 (0.52–1.00) 1 563 5 0.71 (0.57–0.87) 1 625 6 Strength and resistance training 1.80 (0.84–3.87) 64 1 0.77 (0.52–1.14) 120 1 Walking groups NR NR NR 0.95 (0.77–1.18) 222 1 Any exercise interventions NR NR NR Risk of fall-related fracture: 0.34 (0.18–0.63) 810 6 Multicomponent group exercise (combination of 2 or more categories of exercise) 0.71 (0.63–0.82) 3 622 16 0.85 (0.76–0.96) 5 333 22 Multicomponent home-based exercise 0.68 (0.58–0.80) 951 7 0.78 (0.64–0.94) 714 6 Exercise training including only one of gait, balance or functional training 0.72 (0.55–0.94) 519 4 0.81 (0.62–1.97) 453 3 Vitamin D Vitamin D supplementation 1.00 (0.90–1.11) 9 324 7 0.96 (0.89–1.03) 26 747 13 Vitamin D supplementation in people with low vitamin D levels 0.57 (0.37–0.89) 260 2 0.70 (0.56–0.87) 804 4 Home assessment Home safety assessment and modification interventions Overall 0.81 (0.68–0.97) 4 208 6 0.88 (0.80–0.96) 4 051 7 Led by occupational therapist 0.69 (0.55–0.86) 1 443 4 0.79 (0.70–0.91) 1 153 5 Not led by occupational therapist 0.91 (0.75–1.11) 3 075 4 0.94 (0.85–1.05) 2 975 3 Vision Treatment of vision problems 1.57 (1.19–2.06) 616 1 1.54 (1.24–1.91) 616 1 Vision intervention: multifocal to single-lens glasses Rates of falls and outside falls reduced in those who regularly took part in outside activities 597 1 Risk of all falls and outside falls increased in group with little participation in outside activities 597 1 First cataract surgery (women only) 0.66 (0.45–0.95) 306 1 0.95 (0.68–1.33)
Risk of fall-related fracture: 0.33 (0.10–1.05)306 1 Second cataract surgery 0.68 (0.39–1.17) 239 1 1.06 (0.69–1.63) 239 1 Other Multifactorial interventions 0.76 (0.67–0.86) 9 503 19 0.93 (0.86–1.02) 13 617 34 Oral nutritional supplement NR NR NR 0.95 (0.83–1.08) 1 902 3 Gradual withdrawal of psychotropic medication 0.34 (0.16–0.73) 93 1 0.61 (0.32–1.17) 93 1 Medication review and modification 1.01 (0.81–1.25) 186 1 1.03(0.81–1.31) 445 2 Pacemakers (to treat carotid sinus hypersensitivity) 0.73 (0.57–0.93) 349 3 0.78 (0.18–3.39) 171 1 Antislip shoe device for icy conditions 0.42 (0.22–0.78) 109 1 0.56 (0.23–1.38) 40 1 Multifaceted podiatry (foot and ankle exercises) 0.64 (0.45–0.91) 305 1 No reduction 305 1 Cognitive behavioural interventions 1.00 (0.37–2.72) 120 1 1.11 (0.80–1.54) 350 2 Education about fall prevention for patients 0.33 (0.09–1.20) 45 1 0.88 (0.75–1.03) 2 555 4 Professional development for family physicians,† with patient self-assessment, and medication review and modification NR NR NR 0.61 (0.41–0.91) 659 1